Real-time genomic pathogen, resistance, and host range characterization from passive water sampling of wetland ecosystems.

Wetland ecosystems provide interfaces for the transmission of microbial pathogens and antimicrobial resistances (AMR) between migratory birds, wild and domestic animals, and human populations. The efficient surveillance of wetlands is, however, challenging, since the typically low concentration of pathogens requires the sampling of large volumes of water and subsequent targeted detection, which is inherently limited to a few pathogens or AMR genes of interest. Here, we present a holistic, accessible, and cost-efficient framework to characterize the pathogen and resistance load of water sources together with their potential associated hosts by combining passive water sampling through torpedo-shaped devices with nanopore sequencing technology. We used this framework to characterize anthropogenically influenced and natural wetland ecosystems along the East Atlantic Flyway, where we obtained robust assessments of the microbial communities from long-read metagenomic and RNA virome data and showed that anthropogenically impacted wetland ecosystems consistently exhibited higher relative abundances of pathogens and AMR genes. By focusing on avian influenza viruses (AIV), we finally highlight the additional need for targeted screening and whole-genome sequencing of pathogens of interest; we detected and characterized AIV at a third of the monitored sites and used environmental DNA to explore potential animal hosts to better understand the role of wetland ecosystems as One Health interfaces, where the health of animals, humans, and the environment are interconnected and pathogen transmission can occur across these domains. IMPORTANCE: Wetlands connect wildlife, livestock, and people, making them key places to watch for pathogens and antibiotic resistance. Yet potentially harmful microbes are easy to miss in water because they represent only a small fraction of the abundant microbial life in water, making them hard to detect. We paired 3D-printed passive torpedo-shaped samplers with a portable genetic sequencer to analyze all microbes captured. We deployed this approach at 12 wetlands in Germany, France, and Spain. It revealed local microbial communities, identified disease-causing bacteria, and linked many antibiotic resistance genes to likely bacterial hosts. By comparing locations, we observed that sites near cities, farms, or wastewater had higher levels of pathogens and resistance than protected natural sites. Our analysis also recovered all viruses present, including those from mammals, birds, fish, insects, and plants. We also specifically looked for the virus that causes avian flu, found it at several sites, and classified it as low pathogenicity. Because our method is non-invasive to wildlife, affordable, and practical to deploy, it can provide early warnings to conservation and public health agencies and guide action where risks are present.